Two is better than one: A diploid genotype for neural networks

In nature the genotype of many organisms exhibits diploidy, i.e., it includes two copies of every gene. In this paper we describe the results of simulations comparing the behavior of haploid and diploid populations of ecological neural networks living in both fixed and changing environments. We show that diploid genotypes create more variability in fitness in the population than haploid genotypes and buffer better environmental change; as a consequence, if one wants to obtain good results for both average and peak fitness in a single population one should choose a diploid population with an appropriate mutation rate. Some results of our simulations parallel biological findings.

Publication type: 
Articolo
Author or Creator: 
Calabretta, R
Galbiati, R
Nolfi, S
Parisi, D
Publisher: 
Kluwer, Dordrecht , Paesi Bassi
Source: 
Neural processing letters (Dordr., Online) 4 (1996): 149–155. doi:10.1007/BF00426023
info:cnr-pdr/source/autori:Calabretta, R; Galbiati, R; Nolfi, S; Parisi, D/titolo:Two is better than one: A diploid genotype for neural networks/doi:10.1007/BF00426023/rivista:Neural processing letters (Dordr., Online)/anno:1996/pagina_da:149/pagina_a:155
Date: 
1996
Resource Identifier: 
http://www.cnr.it/prodotto/i/296309
https://dx.doi.org/10.1007/BF00426023
info:doi:10.1007/BF00426023
Language: 
Eng